Cancer is caused by cellular defects that produce genetic changes that cause uncontrolled cell proliferation and the loss of the characteristics that define the cell type of origin. As many combinations of genetic changes can cause cancer in the same cell type, cancers that are classified as a cancer type, such as Non-Small Cell Lung Cancer (NSCLC), are really different forms of disease that respond differently to chemotherapy. The hypothesis that this project will test is that the set of mutations that drive a particular typ of cancer (an oncogenotype), will also create weaknesses in that cancer cell type that are not shared with normal cells or cancers with a different oncogenotype. This project will extend the work of a pilot phase CTDD project that has designed methods to identify vulnerabilities in cancers that are caused by their underlying mutational status. Our methods rely upon the use of novel chemical probes and thus are biased towards detecting potentially druggable targets likely to be of therapeutic use. We will screen a large panel of NSCLC and SCLC cell lines with a unique natural products library created at UT Southwestern. In the pilot phase of the CTDD we have screened a subset of this library with genome wide siRNA libraries and 200,000 chemical compounds, allowing us to categorize the lung cancer panel according to vulnerabilities to these reagents. In this current project we will identify natural products that ae toxic to groups of cell lines that are related by the mutations they contain and their sensitivitie to chemotherapeutic agents and siRNA probes. We will identify the mode of action and molecular targets of natural product chemical probes of interest. To do this, we will employ a new bioinformatics tool (FUSION) we have developed to rapidly link the activity of an unknown chemical probe to the activity of siRNAs, miRNAs and compounds with known mode of action. By sequencing the exome of each cell line and comparing mutations found in cell lines to the mutations found in tumors by The Cancer Genome Atlas, we will determine which mutations are linked to each vulnerability, so that this information can be used clinically to identify tumors tht might be treated by targeting the vulnerability. We will validate each vulnerability by determining if the growth of tumor xenografts in mice is antagonized by a chemical probe that targets the vulnerability and/or by delivering siRNAs against the target to xenografts in mice, using nanoparticle delivery. Because we know the response of the cell lines in our panel to known chemotherapeutic drugs, we will be able to report any linkage of an oncogenotype to clinically approved drugs as soon as the functional oncogenotype is defined. Although this project will begin by developing methods to categorize NSCLC and SCLC, these methods can be applied to any cancer for which there is a sufficiently large collection of tumor-derived cell lines. We wil work with the CTDD to apply our tools and natural products library to other types of cancer. At UT SW we have access to a growing panel of ovarian cancers that provide an opportunity for future study.